• Publications
  • Influence
Categorical Reparameterization with Gumbel-Softmax
TLDR
It is shown that the Gumbel-Softmax estimator outperforms state-of-the-art gradient estimators on structured output prediction and unsupervised generative modeling tasks with categorical latent variables, and enables large speedups on semi-supervised classification.
QT-Opt: Scalable Deep Reinforcement Learning for Vision-Based Robotic Manipulation
TLDR
QT-Opt is introduced, a scalable self-supervised vision-based reinforcement learning framework that can leverage over 580k real-world grasp attempts to train a deep neural network Q-function with over 1.2M parameters to perform closed-loop, real- world grasping that generalizes to 96% grasp success on unseen objects.
WAIC, but Why? Generative Ensembles for Robust Anomaly Detection
TLDR
Generative Ensembles is proposed, which robustify density-based OoD detection by way of estimating epistemic uncertainty of the likelihood model, and performs surprisingly well in practice.
Time-Contrastive Networks: Self-Supervised Learning from Video
TLDR
A self-supervised approach for learning representations and robotic behaviors entirely from unlabeled videos recorded from multiple viewpoints is proposed, and it is demonstrated that this representation can be used by a robot to directly mimic human poses without an explicit correspondence, and that it can be use as a reward function within a reinforcement learning algorithm.
Generative Ensembles for Robust Anomaly Detection
TLDR
Generative Ensembles is proposed, a model-independent technique for OoD detection that combines density-based anomaly detection with uncertainty estimation that outperforms ODIN and VIB baselines on image datasets, and achieves comparable performance to a classification model on the Kaggle Credit Fraud dataset.
Sim2Real Viewpoint Invariant Visual Servoing by Recurrent Control
TLDR
This paper trains a deep recurrent controller that can automatically determine which actions move the end-effector of a robotic arm to a desired object and describes how the resulting model can be transferred to a real-world robot by disentangling perception from control and only adapting the visual layers.
Deep Reinforcement Learning for Vision-Based Robotic Grasping: A Simulated Comparative Evaluation of Off-Policy Methods
TLDR
This paper proposes a simulated benchmark for robotic grasping that emphasizes off-policy learning and generalization to unseen objects, and indicates that several simple methods provide a surprisingly strong competitor to popular algorithms such as double Q-learning.
Meta-Learning Requires Meta-Augmentation
TLDR
It is demonstrated that meta-augmentation produces large complementary benefits to recently proposed meta-regularization techniques, and is described as a way to add randomness that discourages the base learner and model from learning trivial solutions that do not generalize to new tasks.
Grasp2Vec: Learning Object Representations from Self-Supervised Grasping
TLDR
This paper studies how to acquire effective object-centric representations for robotic manipulation tasks without human labeling by using autonomous robot interaction with the environment using self-supervised methods.
...
1
2
3
...